Ancient Cuzco

by Brian S. Bauer

  Ice Cores

  Our information on the paleoclimatology and paleoecology of the central Andes comes from various ice and lake-sediment cores of several different research projects. Currently the best-known record of past climatic conditions comes from ice cores extracted from permanent ice caps on the high peaks of the Andes Mountains. The Quelccaya ice cap (Photo 3.1), located at 5,670 m roughly midway between the Cuzco Valley and the Lake Titicaca Basin, is a major repository of environmental data on precipitation, temperature, and dust events at annual resolution for the last 1,500 years (Thompson et al. 1985, 1988; Thompson and Mosley-Thompson 1987; Shimada et al. 1991). In addition, ice cores taken from the Huascaran glacier in the north-central highlands of Peru have yielded important information on past climatic conditions dating as far back as fifteen thousand years (Thompson et al. 1995). By tracing the rate of ice accumulation, changes in oxygen isotope ratios, and the amount of dust particles deposited in the ice over time, these continuous records provide critical data to model the past climate of the Andes. Records such as these, especially those from the Quelccaya ice cores, will be referred to in this study for assessing climate change within the Cuzco region through time. Because these ice cores have been extracted from deposits at extremely high altitudes in remote areas of the Andes, we assume that many of the changes they record reflect broad regional environmental variability, rather than human-induced changes brought about by the activities of local societies.1

  Lake-Sediment Cores

  Other substantial suites of data concerning the paleoecology of the central Andes have been derived from sediment cores extracted from various lakes. The pollen, macrofossils, phytoliths, charcoal, and sedimentological records in these cores can act as a proxy for human impact as well as vegetation and climatic changes. Undoubtedly, the most important lake cores currently available for reconstructing the paleoecology and paleoclimatology of the Cuzco region have been extracted from Lake Marcacocha in the area of Ollantaytambo. Because this small lake forms the basis for much of our climate reconstruction and our understanding of the impact of human activities on the environment in the Cuzco region, it is important to describe it and the sediment cores in some detail.2

  PHOTO 3.1. The Quelccaya ice cap provides a remarkable record of climatic change in the Andes over the past 1,500 years. (Courtesy of Lonnie G. Thompson)

  MARCACOCHA

  Marcacocha is a small, recently infilled lake at an altitude of 3,355 m and currently about 40 meters in diameter (Photo 3.2). It is situated 12 kilometers from the town of Ollantaytambo, where the Patacancha River joins the Urubamba Valley.3 Having most likely formed in the Late Pleistocene, the lake basin is surrounded by Inca and pre-Inca terraces and lies near the agricultural boundary for maize and potato cultivation. The nearby valley slopes contain a number of archaeological sites, the earliest of which dates to ca. 800 BC (Kendall 1992; Early 1995).

  In 1993 two overlapping series of cores, reaching a maximum depth of 8.25 m, were taken from near the center of the lake. The lowest 2 m of these sequences contained well-rounded gravels of fluviatile origin, which were almost barren of organic remains; in contrast, the upper 6 m were rich in organic sediments. These were sampled for pollen analysis to reconstruct the vegetation history of the area. Microcharcoal content (Clark 1982) as well as the ratio of organic to inorganic material were evaluated to understand the burning and erosional history of the catchment.4 Five bulk radiocarbon dates were taken at regular intervals down the six organic meters of the cores, which gave an internally consistent chronology. The oldest of these samples yielded a calibrated date of around 2200 BC (Chepstow-Lusty et al. 1997: 129). In addition, an inorganic horizon was identified at 50 cm, separating the upper peats from the lower lake muds. This was deposited at the time of infilling ca. AD 1960, according to local sources. Hence, this extra time horizon markedly improves the chronology above the topmost radiocarbon date of ca. AD 1400.

  The Marcacocha lake cores, which cover the last 4,000 years, provide the first proxy record of vegetation change from the Cuzco region. Most of the pollen is considered to be from plants that were within the immediate vicinity of the lake. Using the Marcacocha data to compare with events recorded in other lake-sediment and ice cores from the Andes, we can begin to reconstruct the late prehistoric environmental history of the Inca heartland. It should be noted, however, that we do not expect a one-to-one correlation between the events recorded in the lake-sediment cores and those documented in the ice cores. Paleoecological records from small lakes at lower altitudes, such as Lake Marcacocha, are generally assumed to register mostly local events.5

  The Cuzco Environment and Human Impact: 10,000–2000 BC

  Environmental evidence for the Early and Middle Holocene Epochs has not yet been recovered from the Cuzco region. This is unfortunate, since it was during the Middle Holocene, perhaps sometime between 7000 and 5000 BC that hunter-gatherers first arrived in the Cuzco Valley (Chapter 4).6 Nevertheless, it seems that throughout the central Andes this may have been a time of drier conditions. Lake basins in Bolivia and northern Chile support the evidence for drier conditions during the Middle Holocene, though the timing of this subregional hydrological response varies (Abbott et al. 1997; Schwalb et al. 1999; Abbott et al. 2000; Cross et al. 2000; Baker et al. 2001).

  PHOTO 3.2. Overview of the Patacancha Valley showing Lake Marcacocha surrounded by Inca and pre-Inca terraces (Photograph by Alex Chepstow-Lusty)

  By around 3000 to 2000 BC, the climate in the Andean highlands was becoming not unlike that of the modern day. In Peru, the coasts became much drier and the highlands started getting more regular annual rainfall. It may also be during this interval that El Niño events began (Sandweiss et al. 1996). This had major implications for people and may even be the time when agriculture began to be firmly established in the Andes. This would be consistent with the pollen data examined from both Lake Marcacocha (Chepstow-Lusty et al. 1998) and Lake Paca, as well as macrofossil evidence also from the Junén area, which indicates cultivation beginning around this time (Pearsall 1980, 1983; Hansen et al. 1994).

  The Cuzco Environment and Human Impact: 2000 BC–AD 100

  Since the Lake Marcacocha cores provide the best record of past climate change in the Cuzco region during the Late Holocene (Figure 3.1), their temporal divisions are used here (Chepstow-Lusty et al. 1998). These divisions include the following: 2000 BC–AD 100; AD 100–1100; and AD 1100–1993.

  The Marcacocha cores suggest that even before 2000 BC the forests that covered the upper slopes of the Patacancha Valley had already been cleared7 or had never fully recovered from the Early to Middle Holocene period of sustained aridity experienced before human impact began. The charcoal record indicates that the landscape was subjected to regular burning during most of this time interval. This burning, which is still practiced today, was probably done to maintain the soil fertility for agriculture, as well as the quality of pasture for herds of llamas and alpacas.

  FIGURE 3.1. Selected results from the Marcacocha cores. The shaded areas show periods of possible aridity.

  Pollen types indicate that local crops included Chenopodium quinoa, in the family Chenopodiaceae, confirming that cultivation was taking place in the Cuzco region at least as early as 2200 BC.8 The significant occurrence of pollen from Ambrosia arborescens is an indicator of soil disturbance9 and suggests that agricultural terraces were poorly developed during this period. The first ephemeral terrace works, which may have been used during this period, would have been erased when stone terraces were constructed in later prehistory.

  Around 900 BC there is a marked increase in sedges (Cyperaceae), whose increased presence may reflect shrinking (i.e., drier) lake conditions. This is followed by a significant inorganic peak (Chepstow-Lusty et al. 2002), accompanied by a major permanent decline in Ambrosia toward 700 BC. These events may be linked to an abrupt climatic change around 850–760 BC that has been noted in other archaeological and paleoecological studies conducted el
sewhere in the world (e.g., Van Geel et al. 1996). A second phase of sedges (Cyperaceae) centered on 500 BC corresponds with a peak of inorganic material, as does a third phase developing between 10 BC and AD 100. These phases may reflect general drought conditions for the region during these periods (Chepstow-Lusty et al. 2002).

  The production of Chenopodiaceae crops may have reached its peak around 800 BC at the same time that Ambrosia was experiencing a rapid decline. Droughts may well have been superimposed on what was generally a long-term decline in temperature. Shortly after 800 BC, the Chenopodiaceae also experienced a massive decline. Subsequently, there appear to have been minor surges of Chenopodiaceae-oriented agriculture centered on 350 BC and between about 10 BC and AD 100.

  Maize, the most important crop of the Cuzco region today, is observed for the first time in the Marcacocha sedimentary deposits at around 600 BC. Its presence is noted throughout the rest of the sequence, albeit in a seemingly erratic fashion. The presence of maize at Lake Marcacocha is especially noteworthy because this is approximately the altitudinal limit of maize cultivation in the Patacancha Valley today. This may make the lake an especially sensitive repository for climatic information affecting the altitudinal distribution for maize in the region. Nevertheless, because maize’s relatively large-sized pollen is poorly dispersed and, therefore, under-represented in cores, additional work is needed to establish its antiquity and continuity through the Marcacocha record.

  The timing of potato cultivation in the Cuzco region and elsewhere in the Andes remains open to debate. Since most tuberous families in the Andes are insect pollinated, tangible remains of these plants are rarely observed in the pollen record. Potato remains are also more difficult to detect archaeologically than quinoa or maize, since their soft celluloid structure tends to preserve poorly, even after being burned. Excavations in Chile at the Late Pleistocene site of Monte Verde (Dillehay 1989) and at the Early Holocene site of Guitarrero Cave (Lynch 1980) indicate that hunter-gatherers collected wild tubers from an early date. Nevertheless, their importance in prehistoric diets and the timing of their domestication are still being researched.

  The Cuzco Environment and Human Impact: AD 100–1100

  During this period, evidence for Chenopodiaceae-oriented agriculture in the Marcacocha region sharply declines. Quinoa returns in low abundance at the end of this interval, possibly amongst a number of crops, including maize, but it never again reaches the high levels of cultivation experienced in the pre–AD 100 interval. The low proportion of Chenopodiaceae and Ambrosia suggests that temperatures were suppressed during much of this period, and it is possible that agriculture shifted to the production of hardier tuberous crops as well as pastoralism.

  It is important to note that within the Marcacocha record a distinct Cyperaceae (sedge) event, currently thought to reflect a dry period, is centered on AD 550 (Figure 3.1). This is the largest and most defined peak within the Cyperaceae record, although it does not correspond with any major inorganic peak (Chepstow-Lusty et al. 2002). Meanwhile, in the Quelccaya ice core there is a decrease in the ice accumulation record (a proxy for reduced precipitation) and an increasing abundance of dust particles between AD 540 and 600 (Thompson et al. 1985, 1988, 1992).

  The distinct AD 550 Cyperaceae (sedge) peak in the Marcacocha sediment cores and the concurrent indications of a regional decline in precipitation and increase in dust in the Quelccaya ice cores appear to reflect a period of major climatic disruption during the latter half of the sixth century AD. These climatic episodes may have had significant effects on the coastal cultures of Peru. It has been suggested that a series of droughts, as well as several El Niños, were major factors in the dramatic collapse of the Moche polity between AD 560 and 600 on the northern Peruvian coast (Shimada et al. 1991; Thompson et al. 1992). Although the effects of these climatic events on the highland cultures are still not well understood (Paulson 1976; Isbell 1978), it is worth noting that it was during this period that the Wari state expanded from the Ayacucho region of Peru and began a centuries-long occupation of the Lucre Basin near Cuzco (see Chapter 7).

  The Quelccaya ice cores record that a second period of major droughts occurred between AD 1000 and 1100. Supporting evidence for this prolonged period of aridity comes from hiatuses and other sedimentological changes, indicative of low lake levels, in the Lake Titicaca cores. Kolata (1996) and his associates (Binford et al. 1997; Kolata et al. 2000) have proposed that these climatic changes resulted in the large-scale abandonment of the raised-field systems that surrounded Lake Titicaca, which supported the Tiwanaku state (Binford et al. 1996, 1997; Abbott et al. 1997). It is suggested that the Tiwanaku became overly dependent on their raised-field systems and that the state collapsed as a result of these droughts (Ortloff and Kolata 1993; Kolata and Ortloff 1996).

  Unlike the AD 550–600 drought, the prolonged drought of AD 1000–1100 is less clearly recorded in the Marcacocha cores. Even with minor problems in the Marcacocha chronology caused by assuming a constant sedimentation rate between the two top radiocarbon dates (through an interval including rapid inorganic deposition), the Marcacocha cores reflect an increased sedge abundance from about AD 900 onward. This is well before the Quelccaya record registers a marked reduction in precipitation. The two records may be recording different environmental information at this time, as the major declines in precipitation in the Quelccaya ice core occur during the post–AD 1000 era.

  The Cuzco Environment and Human Impact: AD 1100–1490

  Numerous studies have suggested that globally there was an increase in temperature during the first few centuries of the second millennium AD. This increase, called the Medieval Warm Period (ca. AD 1100–1490), is clearly marked by a period of reduced precipitation in the Quelccaya record. The establishment and dramatic success of the tree Alnus acuminata (aliso in Spanish) throughout the Medieval Warm Period may mark the warming of the climate in the Cuzco region (Chepstow-Lusty and Winfield 2000). From the decrease in grasses, it may also be suggested that the slopes of the Patacancha Valley became too valuable to be used for camelid pasture during this period. It is plausible that many of the slopes in the valley were first formally terraced at this time and that the construction of irrigation canals was begun then as well. These fluctuations in climate and environmental resources are currently under study. They are of special interest to archaeologists working in the Cuzco region, since it was during this period that the Inca state (ca. AD 1200–1400) developed (see Chapter 8).

  Toward the end of the Medieval Warm Period there is a significant dust peak centered on AD 1450 in the Quelccaya record. Around this same time, there is also a distinct inorganic layer at Marcacocha. In both cores, these are the largest events of their kind of the last six hundred years; however, their cause(s) is still undetermined. These phenomena are also important to understand, since it was in the early to mid-1400s that the Inca expanded beyond the Cuzco region.

  The Cuzco Environment and Human Impact during the Little Ice Age: AD 1490–1880

  The existence of an appreciably colder period dating from the late fifteenth to the late nineteenth century, known as the Little Ice Age, is now widely recognized by scholars. The Little Ice Age has been identified in the Quelccaya ice cores by a decrease in the oxygen isotope values (from AD 1520 to 1900), an increase in the dust content (from AD 1490 to 1880), and an initial increase in precipitation (from AD 1500 to 1700)10 followed by a dry period (AD 1720–1860; Thompson et al. 1986). The ice cores also appear to record a series of strong El Niño events across the Little Ice Age.

  Both the beginning and the end of the Little Ice Age were very abrupt, as indicated by distinct and dramatic increases in the climate indices for the ice cores. The end of the Little Ice Age is placed at AD 1880, with the modern climate characterized by increased annual mean temperatures (Thompson and Mosley-Thompson 1987: 107). It is also suggested that although the onset of this period began around AD 1490, the most extreme effects did not begin for several centuries (
Thompson and Mosley-Thompson 1987: 105). Thus, it was during the initial, comparatively milder decades of the Little Ice Age that the Inca Empire grew to its maximum size and came to control most of western South America.

  Summary and Discussion

  Although the study of past climate in the Cuzco region and concomitant anthropogenic modifications of the landscape is still just beginning, several important observations can be made. The first hunter-gatherers arrived in the Cuzco Valley during the Middle Holocene. At this time it is likely that conditions drier than exist today prevailed. By 2000 BC, precipitation had increased and the climate conditions resembled those of the modern day. During this period we may see the widespread cultivation of Chenopodiaceae-oriented agriculture, and settlements began to become larger and more permanent.

  The Marcacocha sediment cores suggest that although Chenopodiaceae production may have dominated local agriculture for more than a millennium, it suddenly declined around 700 BC. About this same time, the first maize pollen appears in the lake cores. It is possible, although additional research needs to be conducted, that these events document the arrival of maize in the Cuzco region and a dramatic shift in agricultural practices that would have accompanied it.